Merge pull request #38 from GeneDx/bugfix/requirements

fix pip install

.github/workflows/pythonpackage.yml

@@ -21,9 +21,8 @@ jobs:
       run: |
         python -m pip install --upgrade pip
         pip install pipenv
-        python setup.py develop
+        pip install .
 
-        
     - name: Test with unittest
       run: |
         pip install pytest

setup.py

@@ -31,9 +31,6 @@
         'negspacy',
         'networkx',
         'gensim',
-        'en_core_sci_sm @ https://s3-us-west-2.amazonaws.com/ai2-s2-scispacy/releases/v0.2.4/en_core_sci_sm-0.2.4.tar.gz',
 
-
-    ],
-    dependency_links=['https://s3-us-west-2.amazonaws.com/ai2-s2-scispacy/releases/v0.2.4/en_core_sci_sm-0.2.4.tar.gz#egg=en_core_sci_sm']
+    ]
 )

txt2hpo/nlp.py

@@ -1,24 +1,29 @@
 import spacy
-import en_core_sci_sm
 from negspacy.negation import Negex
 from gensim.parsing.preprocessing import remove_stopwords
 from txt2hpo.config import logger
-from txt2hpo.util import hpo_network
+from txt2hpo.util import hpo_network, download_model
 from nltk.stem import RegexpStemmer
 from spacy.tokens import Token
 
 
 def nlp_model(negation_language="en"):
     try:
+        import en_core_sci_sm
         nlp = en_core_sci_sm.load(disable=["tagger", "parser"])
         nlp.add_pipe(nlp.create_pipe('sentencizer'))
         negex = Negex(nlp, language=negation_language, chunk_prefix=["no"])
         nlp.add_pipe(negex, last=True)
         Token.set_extension('negex', default=False, force=True)
 
-    except OSError:
-        nlp = None
-        logger.info('Negation model could not be loaded\n')
+    except ModuleNotFoundError:
+        rl = download_model("https://s3-us-west-2.amazonaws.com/ai2-s2-scispacy/releases/v0.2.4/en_core_sci_sm-0.2.4.tar.gz")
+        if rl == 0:
+            import en_core_sci_sm
+            nlp = en_core_sci_sm.load(disable=["tagger", "parser"])
+        else:
+            logger.info('Negation model could not be loaded\n')
+            nlp = None
 
     if nlp:
         for not_a_stop in remove_from_stops.split(" "):
@@ -28,15 +33,23 @@ def nlp_model(negation_language="en"):
     return nlp
 
 try:
+    import en_core_sci_sm
     nlp_sans_ner = en_core_sci_sm.load(disable=["tagger", "parser", "ner"])
-    logger.info('Using sci spacy language model\n')
+    logger.info('Using scispaCy language model\n')
 
-except OSError as e:
-    logger.info('Sci spacy language model could not be loaded\n')
-    logger.info('Performing a one-time download of an English language model\n')
-    from spacy.cli import download
-    download('en_core_web_sm')
-    nlp_sans_ner = spacy.load("en_core_web_sm", disable=["tagger", "parser", "ner"])
+except ModuleNotFoundError:
+    rl = download_model(
+        "https://s3-us-west-2.amazonaws.com/ai2-s2-scispacy/releases/v0.2.4/en_core_sci_sm-0.2.4.tar.gz")
+    if rl == 0:
+        import en_core_sci_sm
+        nlp_sans_ner = en_core_sci_sm.load(disable=["tagger", "parser", "ner"])
+        logger.info('Using scispaCy language model\n')
+    else:
+        logger.info('scispaCy language model could not be loaded\n')
+        logger.info('Performing a one-time download of an English language model\n')
+        from spacy.cli import download
+        download('en_core_web_sm')
+        nlp_sans_ner = spacy.load("en_core_web_sm", disable=["tagger", "parser", "ner"])
 
 # these are used in hpo as part of phenotype definition, should block from filtering
 remove_from_stops = "first second third fourth fifth under over front back behind ca above below without no not "

txt2hpo/util.py

@@ -4,6 +4,10 @@
 from phenopy.config import config as phenopy_config
 from phenopy import generate_annotated_hpo_network
 
+import sys
+import subprocess
+import os
+
 
 obo_file = phenopy_config.get('hpo', 'obo_file')
 
@@ -57,3 +61,12 @@ def remove_key(dict_list, key):
         if key in d:
             del d[key]
     return dict_list
+
+
+def download_model(filename, user_pip_args=None):
+    download_url = filename
+    pip_args = ["--no-cache-dir"]
+    if user_pip_args:
+        pip_args.extend(user_pip_args)
+    cmd = [sys.executable, "-m", "pip", "install"] + pip_args + [download_url]
+    return subprocess.call(cmd, env=os.environ.copy())